{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VVBYJTCUTSWUX3LUCOBIJ2VPFL","short_pith_number":"pith:VVBYJTCU","schema_version":"1.0","canonical_sha256":"ad4384cc549cad4bed74138284eaaf2ac7c4d790b0ea453da4d5e819c74f8d8c","source":{"kind":"arxiv","id":"2412.01778","version":1},"attestation_state":"computed","paper":{"title":"HackSynth: LLM Agent and Evaluation Framework for Autonomous Penetration Testing","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Andr\\'as Luk\\'acs, David Imolai, Lajos Muzsai","submitted_at":"2024-12-02T18:28:18Z","abstract_excerpt":"We introduce HackSynth, a novel Large Language Model (LLM)-based agent capable of autonomous penetration testing. HackSynth's dual-module architecture includes a Planner and a Summarizer, which enable it to generate commands and process feedback iteratively. To benchmark HackSynth, we propose two new Capture The Flag (CTF)-based benchmark sets utilizing the popular platforms PicoCTF and OverTheWire. These benchmarks include two hundred challenges across diverse domains and difficulties, providing a standardized framework for evaluating LLM-based penetration testing agents. Based on these bench"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2412.01778","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2024-12-02T18:28:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"064c861f4f5f6ee9ebc47165389a0122ae81c4386f28fcab158fc45ade7762e4","abstract_canon_sha256":"843db3d05659d325314ad6498d61b32666720d371ee873208f05e50d8f5d8d4c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:43:19.472608Z","signature_b64":"UN8cNpITFez2TbYEPtJqyHC8FzzhYcIE2FCtPnUnisJieRXZxsydKCT+RIKDJHJV+Bi9vOq+F+CJQYDqRE+qDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad4384cc549cad4bed74138284eaaf2ac7c4d790b0ea453da4d5e819c74f8d8c","last_reissued_at":"2026-07-05T09:43:19.472075Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:43:19.472075Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HackSynth: LLM Agent and Evaluation Framework for Autonomous Penetration Testing","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Andr\\'as Luk\\'acs, David Imolai, Lajos Muzsai","submitted_at":"2024-12-02T18:28:18Z","abstract_excerpt":"We introduce HackSynth, a novel Large Language Model (LLM)-based agent capable of autonomous penetration testing. HackSynth's dual-module architecture includes a Planner and a Summarizer, which enable it to generate commands and process feedback iteratively. To benchmark HackSynth, we propose two new Capture The Flag (CTF)-based benchmark sets utilizing the popular platforms PicoCTF and OverTheWire. These benchmarks include two hundred challenges across diverse domains and difficulties, providing a standardized framework for evaluating LLM-based penetration testing agents. Based on these bench"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01778","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2412.01778/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2412.01778","created_at":"2026-07-05T09:43:19.472148+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.01778v1","created_at":"2026-07-05T09:43:19.472148+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01778","created_at":"2026-07-05T09:43:19.472148+00:00"},{"alias_kind":"pith_short_12","alias_value":"VVBYJTCUTSWU","created_at":"2026-07-05T09:43:19.472148+00:00"},{"alias_kind":"pith_short_16","alias_value":"VVBYJTCUTSWUX3LU","created_at":"2026-07-05T09:43:19.472148+00:00"},{"alias_kind":"pith_short_8","alias_value":"VVBYJTCU","created_at":"2026-07-05T09:43:19.472148+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07158","citing_title":"Synthetic APTs: the Collapse of TTP-Based Attribution","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05567","citing_title":"ZERO-APT: A Closed-Loop Adversarial Framework for LLM-Driven Automated Penetration Testing under Intelligent Defense","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23243","citing_title":"Are Frontier LLMs Ready for Cybersecurity? Evidence for Vertical Foundation Models from Dual-Mode Vulnerability Benchmarks","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30096","citing_title":"How Reliable Are AI Attackers Against a Fixed Vulnerable Target? A 400-Run Empirical Study of LLM Penetration Testing Consistency","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10749","citing_title":"Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation","ref_index":125,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23243","citing_title":"Are Frontier LLMs Ready for Cybersecurity? Evidence for Vertical Foundation Models from Dual-Mode Vulnerability Benchmarks","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17075","citing_title":"A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17416","citing_title":"Benchmarking Mythos-Linked Bug Rediscovery","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15503","citing_title":"uGen: An Agentic Framework for Generating Microarchitectural Attack PoCs","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2509.06921","citing_title":"Neuro-Symbolic AI for Cybersecurity: State of the Art, Challenges, and Opportunities","ref_index":152,"is_internal_anchor":false},{"citing_arxiv_id":"2509.13021","citing_title":"xOffense: An Autonomous Multi-Agent Framework for Penetration Testing with Domain-Adapted Large Language Models","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10834","citing_title":"From Controlled to the Wild: Evaluation of Pentesting Agents for the Real-World","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24184","citing_title":"Dynamic Cyber Ranges","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05719","citing_title":"Hackers or Hallucinators? A Comprehensive Analysis of LLM-Based Automated Penetration Testing","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14317","citing_title":"Challenges and Future Directions in Agentic Reverse Engineering Systems","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL","json":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL.json","graph_json":"https://pith.science/api/pith-number/VVBYJTCUTSWUX3LUCOBIJ2VPFL/graph.json","events_json":"https://pith.science/api/pith-number/VVBYJTCUTSWUX3LUCOBIJ2VPFL/events.json","paper":"https://pith.science/paper/VVBYJTCU"},"agent_actions":{"view_html":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL","download_json":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL.json","view_paper":"https://pith.science/paper/VVBYJTCU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.01778&json=true","fetch_graph":"https://pith.science/api/pith-number/VVBYJTCUTSWUX3LUCOBIJ2VPFL/graph.json","fetch_events":"https://pith.science/api/pith-number/VVBYJTCUTSWUX3LUCOBIJ2VPFL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL/action/storage_attestation","attest_author":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL/action/author_attestation","sign_citation":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL/action/citation_signature","submit_replication":"https://pith.science/pith/VVBYJTCUTSWUX3LUCOBIJ2VPFL/action/replication_record"}},"created_at":"2026-07-05T09:43:19.472148+00:00","updated_at":"2026-07-05T09:43:19.472148+00:00"}